Well, that's how I understood the comment about listening to people sing on the subway -- she casts it as though the uninitiated listener isn't going to be able to pick out the harmonic context of what the people are singing and will therefore find it semantically void. That definitely happens, but in most cases it's the result of bad singing and not lack of context. Most pop music lines are designed to project a tonal center pretty well, if only by the relatively crude means of just harping on the tonic and related chord tones rather than strictly cadencing, so I think even your average non-musician would have a good sense of the key if a single person with decent pitch were singing such a line in public. But I think we agree on that already!
As for the composer who said that, I think it was Walter Piston? It's definitely an interesting and arguable assertion, especially because it's theoretically testable. I think we probably currently lack the neurological understanding to reliably gauge by any means what tonal context a listener is holding in their mind, but to this layman it seems like a surmountable challenge.
Interesting - I thought she was talking about bad signing making sense to people who knew what the tune was supposed to be, but maybe not. Things fly by pretty fast in her videos; there's often not time to digest her point before she goes on to something else.
I'm not a neuroscientist, but I know that we've had a tonotopic map of the brain for many years: a mapping of frequencies heard to locations in the brain, that can be constructed for any individual. The points lie on a neat curve. And since, often, thinking about a stimulus lights up the same location in the brain associated with perceiving the stimulus, we could conceivably play notes and see what other notes the listener associates with them.
As for the composer who said that, I think it was Walter Piston? It's definitely an interesting and arguable assertion, especially because it's theoretically testable. I think we probably currently lack the neurological understanding to reliably gauge by any means what tonal context a listener is holding in their mind, but to this layman it seems like a surmountable challenge.